Predictive Energy Management for Microgrid Using Multi-Agent Deep Deterministic Policy Gradient With Random Sampling.
Niphon KaewdornhanRongrit ChatthawornPublished in: IEEE Access (2024)
Keyphrases
- random sampling
- energy management
- policy gradient
- multi agent
- single agent
- reinforcement learning
- active learning
- power consumption
- sample size
- energy saving
- partially observable markov decision processes
- multiple agents
- function approximation
- multi agent systems
- energy efficiency
- electric vehicles
- reinforcement learning algorithms
- variance reduction
- optimal control
- power saving
- gradient method
- smart home
- random projections
- sliding window
- markov decision processes
- reinforcement learning methods
- dynamic environments
- path finding
- objective function
- decision problems
- optimal policy
- artificial neural networks
- neural network